17697160. METHOD AND APPARATUS WITH OBJECT CLASSIFICATION simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

From WikiPatents
Jump to navigation Jump to search

METHOD AND APPARATUS WITH OBJECT CLASSIFICATION

Organization Name

SAMSUNG ELECTRONICS CO., LTD.

Inventor(s)

Sangil Jung of Yongin-si (KR)

Seungin Park of Yongin-si (KR)

Byung In Yoo of Seoul (KR)

METHOD AND APPARATUS WITH OBJECT CLASSIFICATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 17697160 titled 'METHOD AND APPARATUS WITH OBJECT CLASSIFICATION

Simplified Explanation

The abstract describes a method and apparatus for classifying objects using a neural network. Here is a simplified explanation of the abstract:

  • The method starts by receiving an input image.
  • The first feature extraction layer of a neural network extracts first feature data from the input image.
  • The second feature extraction layer, which is an upper layer of the first feature extraction layer, provides second feature data.
  • The first and second feature data are merged to generate merged feature data.
  • The object in the input image is then classified based on the merged feature data.

Potential applications of this technology:

  • Object recognition and classification in computer vision systems.
  • Automated image analysis for various industries, such as healthcare, manufacturing, and surveillance.
  • Enhancing the capabilities of autonomous vehicles for object detection and identification.

Problems solved by this technology:

  • Efficient and accurate object classification by utilizing the features extracted from multiple layers of a neural network.
  • Overcoming limitations of traditional object recognition methods by leveraging the power of deep learning and neural networks.

Benefits of this technology:

  • Improved accuracy in object classification due to the combination of features from different layers of the neural network.
  • Enhanced efficiency in processing large amounts of image data.
  • Potential for real-time object classification in various applications.


Original Abstract Submitted

An object classification method and apparatus are disclosed. The object classification method includes receiving an input image, storing first feature data extracted by a first feature extraction layer of a neural network configured to extract features of the input image, receiving second feature data from a second feature extraction layer which is an upper layer of the first feature extraction layer, generating merged feature data by merging the first feature data and the second feature data, and classifying an object in the input image based on the merged feature data.